change default of --sample_inference from independent to pooled

d4straub · d4straub · commit 3bb9be30cc95 · 2026-07-16T14:36:47.000+02:00
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -14,19 +14,20 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ### `Changed`
 
 - [#1018](https://github.com/nf-core/ampliseq/pull/1018) - Change version to 2.19.0dev (by @d4straub)
-
-| previously        | now                            | comment |
-| ----------------- | ------------------------------ | ------- |
-| --nanopore        | --sequencing_type nanopore     |         |
-| --pacbio          | --sequencing_type pacbio       |         |
-| --iontorrent      | --sequencing_type nanopore     |         |
-| --single_end      | --sequencing_type illumina_se  |         |
-| default           | --sequencing_type illumina_pe  | default |
-|                   | --asv_calling auto             | default |
-| default           | --asv_calling dada2            | new     |
-|                   | --asv_calling savont           | new     |
-| --illumina_pe_its | --illumina_pe_readthrough      |         |
-| --extension       | --input_folder_extensions      |         |
+- [#1027](https://github.com/nf-core/ampliseq/pull/1027) - Changed parameters, including the default of `--sample_inference` from `independent` to `pooled` (by @d4straub)
+
+| previously        | now                           | comment |
+| ----------------- | ----------------------------- | ------- |
+| --nanopore        | --sequencing_type nanopore    |         |
+| --pacbio          | --sequencing_type pacbio      |         |
+| --iontorrent      | --sequencing_type nanopore    |         |
+| --single_end      | --sequencing_type illumina_se |         |
+| default           | --sequencing_type illumina_pe | default |
+|                   | --asv_calling auto            | default |
+| default           | --asv_calling dada2           | new     |
+|                   | --asv_calling savont          | new     |
+| --illumina_pe_its | --illumina_pe_readthrough     |         |
+| --extension       | --input_folder_extensions     |         |
 
 ### `Fixed`
 
diff --git a/conf/test.config b/conf/test.config
@@ -31,6 +31,9 @@ params {
     cut_dada_ref_taxonomy = true
     qiime_ref_taxonomy = "greengenes85"
 
+    // ASV generation
+    sample_inference = "independent"
+
     // ASV postprocessing
     vsearch_cluster = true
     decontam = "decontaminate"
diff --git a/conf/test_doubleprimers.config b/conf/test_doubleprimers.config
@@ -27,6 +27,7 @@ params {
     RV_primer = "GACTACHVGGGTATCTAATCC"
     double_primer = true
     input = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Samplesheet_double_primer.tsv"
+    sample_inference = "independent"
     trunc_qmin = 30
     kraken2_ref_taxonomy = "greengenes"
 
diff --git a/conf/test_failed.config b/conf/test_failed.config
@@ -27,6 +27,7 @@ params {
     RV_primer = "GGACTACNVGGGTWTCTAAT"
     input = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Samplesheet_failed_sample.tsv"
     metadata = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Metadata_failed_sample.tsv"
+    sample_inference = "independent"
     dada_ref_tax_custom = "https://zenodo.org/records/4310151/files/rdp_train_set_18.fa.gz"
     skip_dada_addspecies = true
     cut_dada_ref_taxonomy = true
diff --git a/conf/test_full.config b/conf/test_full.config
@@ -19,6 +19,7 @@ params {
     RV_primer = "GGACTACNVGGGTWTCTAAT"
     input = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Samplesheet_full.tsv"
     metadata = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Metadata_full.tsv"
+    sample_inference = "independent"
     dada_ref_taxonomy = "sbdi-gtdb"
     run_pplace = true
     qiime_ref_taxonomy = "greengenes85"
diff --git a/conf/test_glosed.config b/conf/test_glosed.config
@@ -25,6 +25,7 @@ params {
     FW_primer = "GTGYCAGCMGCCGCGGTAA"
     RV_primer = "GGACTACNVGGGTWTCTAAT"
     input = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Samplesheet.tsv"
+    sample_inference = "independent"
     dada_ref_taxonomy = "glosed"
     skip_qiime = true
 
diff --git a/conf/test_iontorrent.config b/conf/test_iontorrent.config
@@ -29,6 +29,7 @@ params {
     sintax_assign_taxlevels = "Domain,Phylum,Class,Order,Family,Genus,Species"
     input = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Samplesheet_it_SE_ITS.tsv"
     sequencing_type = "iontorrent"
+    sample_inference = "independent"
     max_ee = 5
     skip_qiime = true
 
diff --git a/conf/test_multi.config b/conf/test_multi.config
@@ -23,6 +23,7 @@ params {
     config_profile_description = 'Test dataset for multiple sequencing runs to check pipeline function'
 
     // Input data
+    sample_inference = "independent"
     skip_cutadapt = true
     trunclenf = 200
     trunclenr = 150
diff --git a/conf/test_multiregion.config b/conf/test_multiregion.config
@@ -27,6 +27,7 @@ params {
     metadata = params.pipelines_testdata_base_path + "ampliseq/samplesheets/metadata_multiregion.tsv"
     multiregion = params.pipelines_testdata_base_path + "ampliseq/samplesheets/regions_multiregion.tsv"
     sidle_ref_taxonomy = "greengenes88"
+    sample_inference = "independent"
 
     // Prevent default taxonomic classification
     skip_dada_taxonomy = true
diff --git a/conf/test_novaseq.config b/conf/test_novaseq.config
@@ -25,6 +25,7 @@ params {
     // Input data
     input = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Samplesheet_novaseq.tsv"
     binned_quality = "2,11,25,37"
+    sample_inference = "independent"
 
     // Test consensus merging
     mergepairs_strategy = "consensus"
diff --git a/conf/test_pacbio_its.config b/conf/test_pacbio_its.config
@@ -29,9 +29,11 @@ params {
     input = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Samplesheet_pacbio_ITS.tsv"
     metadata = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Metadata_pacbio_ITS.tsv"
     sequencing_type = "pacbio"
+    sample_inference = "independent"
     max_ee = 12
     cut_its = "full"
     its_extractor = "itsxrust"
+
     // addsh = true
     skip_qiime = true
     sbdiexport = true
diff --git a/conf/test_pplace.config b/conf/test_pplace.config
@@ -27,6 +27,7 @@ params {
     RV_primer = "GGACTACNVGGGTWTCTAAT"
     input = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Samplesheet.tsv"
     metadata = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Metadata.tsv"
+    sample_inference = "independent"
     skip_dada_taxonomy = true
     qiime_ref_taxonomy = "greengenes85"
     filter_ssu = "bac"
diff --git a/conf/test_pplace_sheet.config b/conf/test_pplace_sheet.config
@@ -29,6 +29,9 @@ params {
     skip_dada_taxonomy = true
     qiime_ref_taxonomy = "greengenes85"
 
+    // ASV generation
+    sample_inference = "independent"
+
     // Comparison to expected
     expected_sequences = params.pipelines_testdata_base_path + "ampliseq/expected/expected_sequences.fasta"
     expected_abundances = params.pipelines_testdata_base_path + "ampliseq/expected/expected_abundances.tsv"
diff --git a/conf/test_qiimecustom.config b/conf/test_qiimecustom.config
@@ -26,6 +26,7 @@ params {
     FW_primer = "GTGYCAGCMGCCGCGGTAA"
     RV_primer = "GGACTACNVGGGTWTCTAAT"
     input = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Samplesheet.tsv"
+    sample_inference = "independent"
 
     // Comparison to expected
     expected_sequences = params.pipelines_testdata_base_path + "ampliseq/expected/expected_sequences.fasta"
diff --git a/conf/test_reftaxcustom.config b/conf/test_reftaxcustom.config
@@ -26,6 +26,7 @@ params {
     FW_primer = "GTGYCAGCMGCCGCGGTAA"
     RV_primer = "GGACTACNVGGGTWTCTAAT"
     input = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Samplesheet.tsv"
+    sample_inference = "independent"
 
     // Custom reference taxonomy
     dada_ref_tax_custom = "https://zenodo.org/records/4310151/files/rdp_train_set_18.fa.gz"
diff --git a/conf/test_savont.config b/conf/test_savont.config
@@ -31,9 +31,6 @@ params {
     RV_primer = "CGGTTACCTTGTTACGACTT" // 1492R
     retain_untrimmed = true // because primers are not expected in those nanopore reads (were trimmed during demultiplexing)
 
-    // ASV inference
-    sample_inference = "pooled"
-
     // ASV postprocessing
     filter_ssu = "bac"
     min_len_asv = 1000
diff --git a/conf/test_sintax.config b/conf/test_sintax.config
@@ -28,6 +28,7 @@ params {
     input = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Samplesheet_pacbio_ITS.tsv"
     metadata = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Metadata_pacbio_ITS.tsv"
     sequencing_type = "pacbio"
+    sample_inference = "independent"
     max_ee = 12
     cut_its = "its2"
 
diff --git a/conf/test_vsearch_lca.config b/conf/test_vsearch_lca.config
@@ -27,6 +27,7 @@ params {
     RV_primer = "TCCTGAGGGAAACTTCG"
     input = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Samplesheet_pacbio_ITS.tsv"
     metadata = params.pipelines_testdata_base_path + "ampliseq/samplesheets/Metadata_pacbio_ITS.tsv"
+    sample_inference = "independent"
     sequencing_type = "pacbio"
     max_ee = 12
     cut_its = "its2"
diff --git a/nextflow.config b/nextflow.config
@@ -45,7 +45,7 @@ params {
     min_samples                               = 1
     savont_options                            = "--fl-16s"
     multiple_sequencing_runs                  = false
-    sample_inference                          = "independent"
+    sample_inference                          = "pooled"
     binned_quality                            = null
     ignore_binned_quality                     = false
     illumina_pe_readthrough                   = false
diff --git a/nextflow_schema.json b/nextflow_schema.json
@@ -287,8 +287,8 @@
                 },
                 "sample_inference": {
                     "type": "string",
-                    "default": "independent",
-                    "help_text": "If samples are treated independent (lowest sensitivity and lowest resources), pooled (highest sensitivity and resources) or pseudo-pooled (balance between required resources and sensitivity).",
+                    "default": "pooled",
+                    "help_text": "If samples are treated independent (lowest sensitivity and lowest resources), pooled (highest sensitivity and resources) or [only for DADA2] pseudo-pooled (balance between required resources and sensitivity).",
                     "description": "Mode of sample inference: \"independent\", \"pooled\" or \"pseudo\"",
                     "enum": ["independent", "pooled", "pseudo"]
                 },
